> ## Documentation Index
> Fetch the complete documentation index at: https://docs.archetypeai.app/llms.txt
> Use this file to discover all available pages before exploring further.

> ## Agent Instructions
> Start with /introduction/getting-started. Use the Direct Query API (POST /api/lens/query) with the Newton Fusion model (text, image, and video reasoning) or the Newton Omega encoder (time-series embeddings). ATAI_API_ENDPOINT is the base URL https://platform.u1.archetypeai.app with no version path; every endpoint path begins with /api. Pages whose descriptions are marked (Archived) document the legacy Lens runtime — do not use them for new projects.

# About the Projects API

> An API to group the blueprints for one agent effort, and everything derived from them, into projects.

Requires [version 2.0.0](https://docs.archetypeai.app/release-notes/2.0.x.md#v2-0-0) or later of the Archetype platform.

The Projects API creates, lists, renames, and deletes projects. Projects are served under
`/api/projects`, alongside the [Agents API](https://docs.archetypeai.app/api-reference/agents/about).

## What Projects Are

A project groups, within an organization, the blueprints for one agent effort and everything
derived from them. Each project has a `prj_` id, a name, and an optional description.

* **Blueprints** live in a project when you register them there with `project_id` in [Create
  Blueprint](https://docs.archetypeai.app/api-reference/agents/blueprints/create-blueprint.md), or when you copy one in with
  [Copy Blueprint](https://docs.archetypeai.app/api-reference/agents/blueprints/copy-blueprint.md). A blueprint's project is fixed
  when it is created. To reuse a blueprint in another project, copy it.
* **Evals, optimizations, and packages** made from a blueprint in a project are in that project.
  You don't pass a `project_id` when you create them. A blueprint promoted from an optimization
  trial lands in the run's project too.
* **Agent runs** of a package in a project are in that project.

Each of these resources carries a `project_id` field, which is `null` outside any project. Their
list endpoints take a `project_id` filter: a project's `prj_` id lists that project's resources
only, and `none` lists the resources outside any project. Without the filter, a list holds the
resources outside projects plus those of every project you can see.

## Who Can See a Project

* A project is visible to the user who created it and to the organization's admins.
* A project created with a service-account API key is visible to every service-account key of the
  organization instead of a single user.
* Users with the `viewer` role have read-only access: they can't create, change, or delete projects
  or what they hold.

Attempting to access a project you can't see, including everything in it, results in an HTTP
`404` response, as if it did not exist. A project you can see but not modify answers `403` when
you try to change it or what it holds.

## Who Owns a Project's Jobs and Files

The jobs and output files of an eval, optimization, or agent run in a project belong to the
project's creator, whoever starts the run. This way, the creator can always read the outputs.
In a project created with a service-account API key, they belong to no user. `created_by` on the
run still records who started it.

For the same reason, a run in a project may only read input files that the project's creator can
see. Any other input file returns `400` when the run is created.

## Deleting a Project

Only an empty project can be deleted. Delete the blueprints, evals, optimizations, packages, and
runs it holds first; until then, [Delete Project](https://docs.archetypeai.app/api-reference/projects/delete-project.md) returns
`409`.


This documentation is built and hosted on [Mintlify](https://mintlify.com), a developer documentation platform.